Filmology

Filmology is a movie recommendation web application that provides tailored movie suggestions based on user input. It leverages a combination of a content-based filtering approach and the TMDb API to deliver accurate movie details, including posters and recommendations.

🚀 Features

• Tailored Recommendations: Get movie recommendations based on similar titles.
• Trending Movies: Stay updated with the latest popular films.
• Detailed Movie Insights: Access comprehensive information about a movie, including cast, overview, and poster images.

🛠️ Installation

Prerequisites

Before you can run Filmology, ensure you have the following installed:

• Python 3.x
• Flask
• Pandas
• Scikit-learn
• TMDbv3 API
• CountVectorizer (from sklearn.feature_extraction.text)
• TMDb API Key (available at TMDb)

Setup Instructions

1. Clone the repository:

   git clone https://github.com/Alogyn/filmology.git
   cd filmology

2. Install the required dependencies: If requirements.txt is provided, install dependencies with:

   pip install -r requirements.txt

3. Add your TMDb API key: Update the app.py file with your TMDb API key:

   tmdb.api_key = 'your_tmdb_api_key_here'

4. Run the application:

   python app.py

5. Access the app: Open your browser and navigate to http://127.0.0.1:5000/.

📂 Project Structure

.
├── dataset/
│   └── movies2.csv          # Movie dataset
├── preprocessing/
│   └── preprocessing.ipynb  # Preprocessing notebook
├── static/                  # Static assets (CSS, JS, images)
│   ├── styles/
│   ├── images/
│   └── scripts/
├── templates/               # HTML templates for Flask
│   ├── index.html           # Home page
│   ├── recommendation.html  # Recommendations page
│   └── movie_details.html   # Movie details page
├── app.py                   # Flask backend
├── README.md                # Project documentation
├── requirements.txt         # Python dependencies
└── .env                     # Environment variables (e.g., API keys)

🔍 How It Works

Filmology uses content-based filtering to recommend movies based on their metadata (genre, director, cast, etc.). The system calculates the cosine similarity between movies and returns a list of recommendations based on the user's input.

• Data Preprocessing: The preprocessing steps are outlined in the preprocessing/preprocessing.ipynb file.
• Recommendation Algorithm: The application uses CountVectorizer to process text data and calculate similarity scores using cosine_similarity.

🌐 API Integration

Filmology integrates with the TMDb API to fetch and display movie posters, ratings, and details.

API Endpoints

• / (GET): Renders the homepage.
• /recommendation (POST): Provides movie recommendations based on user input.
• /movie/<title> (GET): Displays detailed information about a specific movie.

📈 Technologies Used

• Backend: Flask (Python)
• Data Processing: Pandas, Scikit-learn
• Frontend: HTML, CSS, JavaScript
• API: TMDb API for movie details

🎬 Demo Video

Watch the demo video of Filmology on YouTube:

👨‍💻 Author

• Name: Mohamed Derfoufi
• GitHub: Alogyn
• Email: mohamed.derfoufi.dev@gmail.com

📝 License

This project is licensed under the MIT License. See the [LICENSE] file for more details.